MongoDB Associate Data Modeler - Fact Sheet¶
Exam Overview¶
Exam Name: MongoDB Associate Data Modeler (the exam page heading reads "MongoDB Data Modeler Exam"; the guide title is "MongoDB Associate Data Modeler Exam Guide") Exam Code: None published Duration: 110 minutes Questions: 75 multiple-choice questions, one correct response each. The guide says there may be more than 75, because a handful of unscored questions can be added. Passing Score: Not published; MongoDB sets it statistically per exam and does not publish the required percentage. Only the overall percentage must be met, not a pass in each domain. Cost: $150 USD (exam page and guide, read 2026-11-07) Valid For: Not stated on the exam page (re-read 2026-11-07); the Certification Program Guide is a gated download and was not read, so validity and retake rules are unverified Delivery: Online, proctored; reference materials are not allowed Languages: English Prerequisites: None Discounts: Completing a Learning Path earns 50% off the exam; students and educators can get a free exam through the student and educator programs (exam page) Extended time: Available for candidates with a disability or for whom English is a second language; email certification@mongodb.com at least 72 hours ahead (guide)
The guide is "Version 2.0". The credential is described as validating "knowledge and skills needed to show competency in data modeling on MongoDB": design, build and evolve effective data models, and incorporate data governance into the work. A score report states pass or fail and the percent correct for each domain.
📖 Exam page - format, fee, time, discounts and study links 📖 Exam study guide - the free guide (a PDF shown in the course viewer) with domains, weights and objectives 📖 MongoDB Data Modeling Path - the recommended learning path 📖 Practice questions - MongoDB's own sample questions (requires a MongoDB University login) 📖 Certification Program Guide - program policies and testing experience
Domains and Weights¶
| # | Domain | Weight |
|---|---|---|
| 1 | Advanced Schema Design Patterns & Anti-patterns | 15% |
| 2 | Fundamentals of Data Transformation | 6% |
| 3 | CRUD Operations | 3% |
| 4 | Relational to Document Model | 12% |
| 5 | Performance Tools and Techniques | 6% |
| 6 | Indexing Design Fundamentals | 15% |
| 7 | Schema Design Optimization | 16% |
| 8 | Monitoring and Evolving Data Models | 27% |
Domains 8, 7, 1 and 6 together carry 73% of the exam.
Target Audience¶
- Developers and data modelers who design, implement or maintain MongoDB data models and application queries in development or production
- People with some experience in applying schema design principles, using indexing strategies and building aggregation pipelines
- Candidates who completed the Data Modeler learning path (strongly encouraged by the guide)
Domain 1: Advanced Schema Design Patterns and Anti-patterns (15%)¶
📖 Schema Design Patterns - the pattern families 📖 Document and Schema Versioning - versioning use cases 📖 Use the Approximation Pattern - fewer writes for approximate values
Key Facts: - Approximation updates a stored value only when it moves by a step (the docs use steps of 100) - Schema Versioning keeps several schema versions in one collection with a schemaVersion field - Document Versioning keeps old versions in a separate collection
Domain 2: Fundamentals of Data Transformation (6%)¶
📖 Aggregation Pipeline - stages feed each other 📖 $unwind - arrays to documents
Key Facts: - Only $merge and $out make aggregate() write - $unwind drops null, missing or empty arrays unless preserveNullAndEmptyArrays is true - $set is an alias for $addFields; $group does not order output
Domain 3: CRUD Operations (3%)¶
📖 MongoDB CRUD Operations - insert, find, update, delete
Key Facts: - Writes are atomic at the level of a single document - updateOne, updateMany, replaceOne, deleteOne, deleteMany
Domain 4: Relational to Document Model (12%)¶
📖 Embedded Data - when to embed 📖 Referenced Data - when to reference 📖 Schema Validation - validation rules
Key Facts: - Documents must be under 16 mebibytes - validationLevel: strict (default), moderate, constraint (constraint is new in 9.0); validationAction: error (default), warn, errorAndLog (8.1) - Workload table: action, type, information, frequency, priority
Domain 5: Performance Tools and Techniques (6%)¶
📖 Database Profiler - levels and slowms 📖 Analyze Query Plan - reading explain output
Key Facts: - Profiler is off by default; default slow threshold is 100 ms; output goes to system.profile - COLLSCAN plus totalKeysExamined: 0 means no index was used
Domain 6: Indexing Design Fundamentals (15%)¶
📖 Index Types - the types 📖 The ESR Guideline - key order 📖 Compound Indexes - prefixes and limits
Key Facts: - ESR: Equality, then Sort, then Range; ERS if the range is very selective - A compound index can hold up to 32 fields and one hashed field - Index prefixes: leading fields only
Domain 7: Schema Design Optimization (16%)¶
📖 Single Collection Pattern - one collection, many types 📖 Group Data - bucket and outlier 📖 Archive Pattern - move old data 📖 Choose a Shard Key - cardinality, frequency, monotonic
Key Facts: - Time series collections apply the bucket pattern automatically - Archive: external storage such as Amazon S3 is usually best - A monotonic shard key sends all inserts to one chunk
Domain 8: Monitoring and Evolving Data Models (27%)¶
📖 Polymorphic Data - inheritance 📖 Store Computed Data - computed pattern 📖 Schema Design Anti-Patterns - the five anti-patterns
Key Facts: - Performance Advisor needs an M10 or higher Atlas cluster; Compass Performance Insights also flags anti-patterns - Unbounded arrays and bloated documents are fixed by Subset and referencing - Extended Reference copies a few rarely changing hot fields
Exam Tips¶
- Domains 8, 7, 1 and 6 are 73% of the exam. Study patterns and indexing first.
- Symptom to pattern. Most questions describe a problem; name the pattern or anti-pattern.
- Know the price of each pattern. The right answer often depends on the trade-off.
- Read the ESR sentence twice. The ERS exception is a likely trap.
- Time: 110 minutes for 75 questions is about 1.4 minutes each.
Quick Reference¶
// Run on MongoDB 8.3.13 (mongo:8 Docker image) on 2026-11-07 where shown in the notes.
db.createCollection("members", {validator: {$jsonSchema: {bsonType: "object", required: ["email"], properties: {email: {bsonType: "string"}}}}});
db.runCommand({collMod: "members", validationLevel: "moderate", validationAction: "warn"});
db.events.createIndex({status: 1, user: 1, ts: 1}); // Equality, Sort, Range
db.events.find({status: "open", ts: {$gte: 5000}}).sort({user: 1}).explain("executionStats");
db.setProfilingLevel(1, {slowms: 100});